AB jailbreaking - a novel hybrid framework for exploitation of adversarial vulnerabilities in LLMs

Abstract Large language models (LLMs) have advanced rapidly but remain vulnerable to adversarial “jailbreaking” attacks that elicit harmful or disallowed outputs. We propose AB-JB, a three-stage hybrid jailbreak framework that combines black-box semantic adversarial prompt variant generation with a...

Descripción completa

Detalles Bibliográficos

Autores principales:	Abrar Ahmad, Mehwish Naseer, Usman Qamar, Monther Alfuraidan
Formato:	Artigo
Lenguaje:	Inglês
Publicado:	Nature Portfolio 2026-04-01
Colección:	Scientific Reports
Materias:	Jailbreaking Adversarial attacks LLMs Hybrid attacks White-box attacks Black-box attacks
Acceso en línea:	https://doi.org/10.1038/s41598-026-44403-w
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!